Varicella-zoster virus (VZV) causes varicella in susceptible individuals, establishes latent infection in neural ganglia and reactivates as herpes zoster. The importance of VZV cellular immunity is illustrated by the severity of varicella in children with T-cell deficiencies. In the studies proposed, new methods employing vaccinia recombinants that express VZV proteins and synthetic peptides that correspond to regions of major VZV proteins will be used analyze the viral protein targets of T- cell immunity during varicella and in longterm naturally acquired or vaccine-induced immunity. Vaccinia recombinants expressing glycoproteins (gp) I, IV and V and immediate early (IE) protein, will provide specific antigens to analyze helper T-cell proliferation and IFN-g production to VZV proteins during acute primary VZV infection and persistent cellular immunity. Helper T-cell epitopes of gp I, IE protein, nucleocapsid proteins (NA) and viral DNA polymerase will de defined with synthetic peptides. Synthetic peptides will help to determine whether a few regions of VZV proteins are predictably recognized by T-cells from individuals of varying HLA types. The kinetics and protein specificity of CTL will be examined during acute varicella using autologous lymphoblastoid cells (LCL) infected with VZV or VZV-vaccinia recombinants as targets. The frequency of memory CTL specific for particular VZV proteins will be assessed b y the limiting dilution cytotoxicity assay. VZV epitopes recognized by CTL will be determined using synthetic peptides to sensitize LCL. The identification of VZV proteins that elicit cellular immunity and of T-cell epitopes for both helper and CTL responses can be used to guide new approaches to VZV immunization. Cell- associated viremia, a critical event during the pathogenesis of primary VZV infection will be investigated further using in situ hybridization with VZV and RNA probes, including probes for immediate early gene 62. Cell-associated viremia in VZV pathogenesis will also be investigated by making viral strains that have been modified by recombinant DNA techniques. Mutant strains of VZV will be developed by inserting E. coli lacZ gene sequences to disrupt VZV gene function recombinant strains will be tested for the effects of insertional mutagenesis upon the capacity to infect PBMC in vitro. The ultimate goal of these experiments will be to understand VZV pathogenesis at the level of gene function e.g., to identify gene sequences associated with the ability of the virus to infect PBMC or other target cell of the virus. While it is clearly a longterm objective, such genetic manipulations of VZV could yield a recombinant VZV strain that induced persistent humoral and cell-mediated immunity to VZV without retaining the capacity to infect PBMC and neural cells.